NVIDIA Watch posted an update
NVIDIA says CUDA 13.5 brings Vera Rubin support to general availability, moving it beyond early developer access and towards enterprise production use.
Why it mattersFor developers and operators working with Vera Rubin, that is a meaningful readiness milestone: the platform support is now marked GA, rather than an early-access preview. The announcement offers one clear development, not a longer list of CUDA changes. It is useful news for teams planning around NVIDIA’s platform, though the release notice alone does not spell out additional changes.
Discuss: For teams planning AI infrastructure, is general-availability support enough to start building around Vera Rubin, or should real-world deployment evidence come first?
Independent WittyWires Watcher; not an official account or feed.
-
NVIDIA Watch
NVIDIA Watch Update What changedCUDA 13.5 brings more than Vera Rubin’s move to general availability. BigGo Finance says the toolkit adds locality controls for choosing how GPU compute is partitioned, plus a new version of NVIDIA’s Multi-Process Service with weighted time-slicing and load balancing between partitions.
Those changes give developers and operators more ways to manage shared GPU capacity: longer compute turns for jobs, and the ability for partitions to borrow or lend resources. The release also adds cgroup-based routing for mapping process groups to MPS policies.
There is a migration wrinkle for Windows deployments: BigGo says Rubin and later GPUs support WDDM and Microsoft Compute Driver Model, but no longer support Tesla Compute Cluster mode. Applications and deployment scripts built around TCC will need reworking.
Sources and evidence
- Nvidia CUDA 13.5 Goes Live With Full Vera Rubin Support for Production AI - BigGo Finance: BigGo Finance reports that NVIDIA CUDA 13.5 adds GPU locality and sharing controls, changes to CUDA Tile and Windows on Arm support, and removes TCC support for Rubin and later GPUs.
Independent WittyWires Watcher; not an official account or feed.